Direct Laves Phase Crystallization in Undercooled W-Nb-Hf-Zr Alloy

Refractory complex concentrated alloys present a difficult solidification problem: very high melting temperatures, strong chemical interactions, and restricted atomic diffusion all influence which phases can form from the liquid state. Their solidification behavior is especially important because the phases selected from the liquid state can determine not only microstructure, but also stability and environmental resistance. To monitor directly these liquid-solid transitions is not simple because refractory liquids require extreme temperatures and are easily affected by crucible reactions and heterogeneous nucleation. Conventional processing can introduce unwanted reactions, heterogeneous nucleation, and complex thermal histories that make it harder to isolate the true competition between solid solutions and intermetallic phases. For alloys containing W, Nb, Hf, and Zr, achieving the liquid state requires extreme high temperature, and multicomponent chemistry restricts how quickly atoms can redistribute during solidification.

Laves phases are especially relevant here because they are ordered intermetallic compounds whose formation is controlled by atomic size, electronic structure, and local packing. In multicomponent refractory alloys, their formation cannot be treated simply as the low-temperature consequence of equilibrium thermodynamics. A Laves phase may be thermodynamically favored, but does not appear when the liquid-solid transition moves too quickly for the required atomic rearrangement. When the liquid is undercooled beyond a critical point, the balance changes: the ordered Laves phase can nucleate directly from the melt instead of waiting for the slower peritectic reaction to form it and the challenge is to determine when stability can actually be expressed during solidification.

In a recently published research paper in Acta Materialia, Mr. Kelun Liu, Dr. Ruilin Xiao, Mr. Bohan Sun, Professor Ying Ruan, and Professor Bingbo Wei from Northwestern Polytechnical University developed an electrostatic-levitation-based solidification approach for resolving the growth mechanism of a refractory multicomponent Laves phase in W₂₅Nb₂₅Hf₂₅Zr₂₅ alloy. They identified the Laves phase as C15-type (W,Nb)₂(Hf,Zr), with W/Nb and Hf/Zr occupying distinct crystallographic sublattices supported by atomic-resolution imaging and first-principles calculations. They established a processing-dependent phase-transition map in which near-equilibrium peritectic formation, non-equilibrium BCC phase selection, and deep-undercooling direct Laves crystallization are separated by undercooling and cooling-rate conditions. They also linked direct Laves formation and altered BCC2 boundary character to improved pitting resistance in sulfuric acid solution.

The research team showed that W₂₅Nb₂₅Hf₂₅Zr₂₅ does not follow a single solidification route and found that below a critical undercooling of about 395 K, the alloy solidified through three sequential body-centered cubic phases. The primary BCC1 phase, enriched in W and therefore associated with the highest melting temperature, formed first from the undercooled liquid. BCC2 then grew epitaxially from BCC1, maintaining a cube-on-cube orientation relationship with a very small average misorientation. A third BCC phase appeared at grain boundaries, enriched in W and Hf relative to the surrounding phases. This sequence is scientifically important because it shows that a multicomponent alloy with strong Laves-forming propensity can still avoid Laves formation when the kinetic path does not permit sufficient solute redistribution.

The authors observed once ΔT exceeded 395 K, the primary phase switched from BCC1 to a Laves phase identified as (W,Nb)₂(Hf,Zr), followed by formation of a BCC2 matrix. The growth kinetics reflected this discontinuity. The BCC1 and Laves regimes were described by different power-law relationships, with the primary Laves phase appearing only after the undercooled liquid crossed the threshold corresponding to the interval between the liquidus and peritectic transition temperature. The researchers also measured the growth velocity of BCC2 as a function of peritectic undercooling and found it to be much faster than the Laves phase growth. The design choice of independently varying undercooling and peritectic undercooling therefore separated the origin of the Laves phase from the later BCC2 growth event, making the phase-selection mechanism clearer rather than treating the final microstructure as a single solidification product. Crystallographic analysis showed no fixed orientation relationship between the Laves phase and BCC2 under these conditions, consistent with independent formation events. The final deeply undercooled microstructure contained faceted Laves particles uniformly embedded in the BCC2 matrix.

The team performed atomic-resolution characterization to determine the structural identity of the intermetallic phase and found that the Laves phase adopted a C15-type cubic structure, with W and Nb occupying the smaller-atom sublattice and Hf and Zr occupying the larger-atom sublattice. First-principles calculations supported this arrangement. The calculated formation energy of (W,Nb)₂(Hf,Zr) was far lower than those of the BCC phases and also lower than alternative atomic substitution arrangements within the Laves structure. This result is central to the paper’s logic: the Laves phase is the most stable phase considered, but its appearance depends on whether solidification conditions allow that stability to be realized.

The investigators conducted near-equilibrium levitation experiments and found at very small undercooling and low cooling rate, the Laves phase appeared wrapped around primary BCC1, consistent with a peritectic transition from liquid plus BCC1 to (W,Nb)2(Hf,Zr), followed by a eutectic reaction involving Laves and BCC2. Under faster non-equilibrium conditions below the critical undercooling, that peritectic reaction was suppressed. The authors attribute this suppression to restricted atomic diffusion in the chemically complex liquid-solid environment, especially for the larger Hf and Zr atoms. A comparison with a simpler Zr-W binary peritectic system strengthened the interpretation: in the binary alloy, the peritectic product persisted under undercooling, whereas the multicomponent W-Nb-Hf-Zr alloy could bypass it entirely. Additionally, deep undercooling studies showed that instead of allowing the usual peritectic pathway to proceed, it supplied enough thermodynamic driving force for direct nucleation of (W,Nb)₂(Hf,Zr) from the liquid.

The team showed in sulfuric acid solution, the sample solidified at higher undercooling showed a pitting potential of 2.11 VSCE, about 40 percent higher than the lower-undercooling condition. Impedance behavior indicated greater charge-transfer resistance and a more capacitive response, while surface analysis showed passive films containing ZrO2, HfO2, WO3, and Nb2O5. The higher-undercooling sample had slightly higher fractions of ZrO2 and HfO2 and a higher lattice oxygen content. Corrosion morphology also changed: pits in the lower-undercooling alloy were associated mainly with BCC3 at grain boundaries, whereas the deeply undercooled alloy showed fewer, shallower pits at Laves/BCC2 interfaces, with the Laves phase itself remaining unaffected.

The engineering importance of the findings of Northwestern Polytechnical University scientists is that solidification pathway control can be used as a practical design variable for refractory alloys. In high-temperature structural materials, especially those based on W, Nb, Hf, and Zr, processing conditions often decide whether the final alloy contains metastable BCC solid solutions, ordered intermetallic phases, or mixed microstructures with very different boundary populations and corrosion responses. By showing that deep undercooling can trigger direct crystallization of a C15-type (W,Nb)₂(Hf,Zr) Laves phase, this work provides a route for engineering microstructures through liquid-solid transitions rather than relying only on post-solidification heat treatment. One immediate application is in the development of refractory complex concentrated alloys for severe thermal and chemical environments. Components exposed to high temperature, acidic media, or aggressive service conditions require not only phase stability but also resistance to localized degradation.

The findings also have relevance for rapid solidification technologies, including containerless processing, laser-based melting andadvanced casting of reactive refractory alloys where steep thermal gradients and non-equilibrium cooling are common. In such processes, deep undercooling and rapid thermal extraction can shift phase selection away from equilibrium pathways. The study gives a mechanistic basis for exploiting that shift deliberately: a diffusion-limited peritectic reaction may be bypassed, while direct intermetallic nucleation becomes possible once the thermodynamic driving force is sufficiently high. The results position undercooling as a controllable phase-selection parameter in refractory alloy processing, especially when diffusion-limited peritectic reactions compete with direct intermetallic nucleation. For refractory alloy development, this could help guide processing windows that avoid undesirable grain-boundary phases, promote stable intermetallic dispersions, refine microstructure, and improve resistance to localized corrosion. The practical value is therefore in linking processing, phase nucleation, atomic structure, and service-relevant behavior within a single design framework.